The Challenge of Reducing the Global Incidence of Civil WarMichael D. IntriligatorEconomics, Political Science, and Policy Studies, UCLA

Perspective Paper

This paper was produced for the Copenhagen Consensus 2004 project.

The final version of this paper can be found in the book, ‘Global Crises, Global Solutions: First Edition’,

edited by Bjørn Lomborg

(Cambridge University Press, 2004)

1

March 30, 2004Opponent Note to the Challenge Paper,

“The Challenge of Reducing the Global Incidence of Civil War” byPaul Collier and Anke Hoeffler as prepared for the Copenhagen Consensus

Michael D. IntriligatorProfessor of Economics, Political Science, and Policy Studies, UCLA

Senior Fellow, The Milken Institute andThe Gorbachev Foundation of North America

1. Purpose of this Opponent Note

The purpose of this opponent note is to provide a counterbalance to thechallenge paper, “The Challenge of Reducing the Global Incidence of Civil War”by Paul Collier and Anke Hoeffler (the “CH Paper”). It will evaluate the issuescovered in the CH paper and will reach some conclusions about these issues,some of which differ from the paper or go beyond it. It will, in addition, discussother issues involved in civil wars as well as issues involved in other types ofconflicts that are not treated in this paper.

2. The Focus of the CH Paper

The focus of the CH paper is civil wars, especially those in Africa, whichhas been the area of concern of its authors. Their paper is largely based on theirrecent book Breaking the Conflict Trap: Civil War and Development Policy (SeeCollier 2003), which, in turn, is based on many past publications, particularlythose of Paul Collier. The authors correctly note that international conflict hasbecome less common while civil wars have become more common. This is true,but international conflicts do, of course, exist and some are of great importance,such as the recent U.S. led wars in Afghanistan and Iraq and their aftermathsthat should also be addressed in a study of conflict and its avoidance.

Civil wars are, nonetheless, extremely important, particularly in Africa,where some of the biggest wars have been fought in recent years and up to thepresent. (See Ali and Matthews, Eds., 1999). Huge civil wars have been foughtor are being fought in, among other places, Angola, Congo, and Sudan.Unfortunately these wars are largely unreported and, if they are reported in thepress they are mostly confined to the back pages of publications while they arelargely unreported in the broadcast media. As a result, these wars are mostlyunknown to the general public, and many specialists have not treated them indepth, with Paul Collier’s work in this area being an important exception. Thepaper concerns civil wars in Africa, such as those mentioned above but alsothose in Liberia, Sierra Leone, Somalia, and Cote d’Ivoire. It also concerns civilwars on other continents, such as those in Asia (including Afghanistan andCambodia), Europe (including Bosnia and Serbia) and Latin America (includingColombia, El Salvador, and Haiti),

Copenhagen Consensus Opponent NoteNot to be released before 23 April 2004

2

3. The Benefits of a Reduction in the Global Incidence of Civil War

The CH paper attempts to establish credible lower-bound estimates of thebenefits that would stem from a reduction in the global incidence of civil war.They note that these benefits accrue at three levels: national, regional, andglobal, although one could also add the local level. At the national level theauthors state that the benefits are partly economic and partly social. One mightadd the political benefits of a stable national government vs. the instability thatwould stem from civil war, with a change in government or many such changesthat could lead to a failed state. There are also other benefits – health,environment, etc., some of which are discussed below. The CH paper measuresthe economic benefits of avoiding civil war by the effects of civil war on growth,and it measures the social benefits in terms of Disability Adjusted Life Years(DALYs). The authors also estimate the regional benefits of avoiding war in termsof its effects on economic growth. Finally, the global benefits of conflict reduction,as the authors correctly note, are the most difficult to estimate and are leftunquantified, although the paper does discuss in general terms the globalimpacts of civil wars in terms of AIDS, drugs, and safe havens for terrorists.

4. Three Opportunities Treated in the CH Paper

The CH paper focuses on three opportunities: the prevention of civil war incurrently peaceful environments, the shortening of conflicts in currently war-ridden environments, and the reduction in the risk of the resumption of conflict inpost-conflict situations. These so-called “opportunities” are analogous toSchelling’s canonical goals of arms control: to prevent war; to reduce thedamage resulting from war, which in a civil war is related to the length of the war;and to reduce the cost of arming, which is related to the war resuming in the civilwar context. (See Schelling, 1960, 1966 and Schelling and Halperin, 1961)

The CH paper states that probably the highest payoff is from the third ofthese opportunities: an improved intervention in post-conflict situations to reducethe risk of the war resuming, but this is questionable. Given the huge costs ofcivil wars in their first days and weeks, as seen, for example, in the genocide inRwanda, a case can be made that the highest payoff is from the first opportunity,simply preventing civil war in the first place. The paper states that renewedviolence accounts for around half of all global civil wars and argues that suchresumptions of conflict provide an opportunity for “highly focused interventions,”while they refer to prevention as a “highly diffuse approach.” In many cases,however, it is possible to foresee the outbreak of violence in a nation and to takesteps to forestall it, while in others it might be able to contain it at an early stage,either of which would yield a very high payoff. Thus, the focus of internationalpublic policy should be on all of these opportunities, with a balanced approach toall three.

3

5. The CH Instruments for Conflict Prevention: Aid, Transparency in NaturalResources, and Some Others

The CH paper investigates two instruments for conflict prevention: aid andtransparency in natural resources. The former is based on their reasoning thatthe risk of conflict is much higher in countries with low per-capita income,negative growth, and dependence on natural resource exports, such as many ofthe nations in Africa that have suffered from civil war. They argue thatinterventions that improve the economic characteristics of nations can reducetheir tendency to engage in civil conflict. Their approach is based on theirassumptions that a country receiving aid will experience higher growth that this,in turn, will reduce the chance of conflict. Both of these assumptions are highlyquestionable, however. In many cases, particularly in sub-Saharan Africa, aidfunds have done little more than add to the offshore bank accounts of dictatorsthrough corruption and other criminal acts. Examples include Mobutu in Zaireand Charles Taylor in Liberia. Such aid does not necessarily add to economicgrowth. This is also shown in the empirical findings of the CH paper, where theyestimate that an extra two percentage points of aid raises the growth rate by only0.2 percentage points, which they admit is a “relatively modest gain” and whichthey attribute, to the problem of diminishing returns. As they note, “…conflictprevention achieved purely by unselective aid programmes to low-incomecountries is not very cost-effective. The benefits amount to less than ten percentof the costs.” This lack of effectiveness of aid in preventing conflict is reiterated intheir conclusions as: “The most disappointing instrument is, in a sense, the mostobvious and the most readily available. This is aid for conflict prevention.”Probably more important than financial aid is technical assistance to help createthe institutions of a modern economy that can facilitate growth and development.Establishing such an infrastructure could have substantial payoffs, as seen in theliterature on economic institutions. (See North, 1990)

As to the second part of the CH reasoning, higher growth does notnecessarily lead to reduced conflict since economic resources may just feed theacquisition of arms and even make the country a more attractive target forexpatriates who would like to take it over and commandeer these resources. Theauthors find in their empirical work that three economic characteristics havesignificant and substantial effects on the risk of conflict: the level of income, itsrate of growth, and the degree of dependence on primary commodity exports, butthe effect of growth appears to be relatively small, a one percent increase in thegrowth rate sustained over a period of ten years having the effect of reducing therisk of civil war over the first five-year period from 13.8 percent to 12.7 percent, arelatively small effect. Taking account of the income level effect lowers the risk inthe second five-year period to 12.2 percent, but even this is still a relatively smalleffect.

The other instrument that the CH paper treats is greater transparency innatural resources, which they see as “an important and cheap practical

4

instrument.” They refer to the adverse effects of natural resource dependence ondevelopment but do not elaborate on why that is the case or how greatertransparency in this area could prevent conflict. One argument might be thatgreater transparency would enable one to trace the use of natural resources tofund a continuing civil war, as in the case of so-called “conflict diamonds.” Theseare diamonds that are exported, where the proceeds of their sale are used tocontinue a civil war, as happened in Angola. If greater transparency could reducethe chance of this occurring then this instrument might be a useful way to preventor shorten a civil war. A counterargument, however, would be that greatertransparency would identify where a country is building up its capital, makingthem the target for a civil war. In any case, the mechanisms connecting greatertransparency in natural resources to conflict prevention must be elaborated. Infact, there are countries that are heavily dependent on exports from extractiveindustries that have not had a civil war and, conversely, there are counties withno such dependence that have had a civil war. Thus, this issue must be studiedmore carefully, looking at the effects of dependence on natural resources oncorruption and the way revenues from sales of such resources have fueled war insome countries but not in others.

The CH paper does not consider other possible instruments to preventconflict. Looking at the issue as an economist, one might treat both the supplyand demand for conflict. As to the supply, conflict in civil wars requires arms, andmost poor countries, such as those in Africa, do not have the indigenouscapability to produce arms. Depriving these nations of arms by cutting off theirarms supply could be an effective means to reduce conflict from breaking out,continuing, or restarting. The CH paper mentions that control of the trade inarmaments as one worth serious effort. Oscar Arias, the former President ofCosta Rica and the winner of the Nobel Peace Prize, has made a proposal alongthese lines. (See Arias, 1996, 2002) His proposal was that the major industrialnations, particularly the permanent members of the UN Security Council, China,France, Russia, the United Kingdom, and the United States, impose an embargoon all arms shipments to sub-Saharan Africa. These are the major armsexporting nations and if they would stop shipping weapons into the region thatcould go a long way to reducing conflict in this part of the world, a region that hassuffered inordinately from civil wars, including some of the largest in recentyears. A type of precedent for this Arias proposal is the Tripartite Agreementunder which three of these major powers, France, the United Kingdom, and theUnited States, agreed to limit the quantity and type of weapons they wereshipping to the Middle East, which worked reasonably well until the Soviet Unionstarted to ship weapons to Egypt in 1958, after which the agreement collapsed.

More generally, one of the best ways to prevent civil war is to reduce theresources of the people who would engage in such conflict, including not onlytheir access to arms but also their access to financial resources, to personnel, tomercenaries, to training, etc. Such a reduced access could be achieved byconcerted actions of neighboring states in policing the borders and by the great

5

powers and international organizations in monitoring of flows of arms, money,and people to the nation in risk. (Sandler, 2002 argues that this is the bestapproach to fighting terrorism)

The other side of this approach is the demand for conflict. This is where thetools of diplomacy could be important in mediating conflicts, establishing a peaceagreement, or installing troops to preserve or restore order. These types ofinstruments for preventing conflict would probably be more effective in preventingconflict than the two that are discussed in the CH paper. (See Stedman,Rothchild, and Cousens, 2002 on establishing peace agreements. See alsoPresident Jimmy Carter, 1984, 2001 about his experience in negotiatingagreements to prevent war)

Another instrument to prevent conflict is the use of peacekeeping forces,whether from the UN or regional military groups, as happened in the case ofLiberia, or even former colonial powers, as happened in the case of SierraLeone. These peacekeepers can restrain the warring parties and preserve thepeace, avoiding conflict, whether before it starts or during a civil war or to preventthe repeat of such a war. (See Klein and Marwah, 1996, Rotberg, et. al., 2000,and Brahimi, 2000)

6. An Instrument for Shortening Conflict: Tracking of Natural Resourcesand Some Others

The CH paper considers only one instrument for shortening conflict: thetracking of natural resources, which they suggest is the most promisingintervention. As their paper notes, such tracking through a certification processhas the potential of choking off those arms that are purchased with sales of oil,gold, diamonds, and other natural resources. An example is the Kimberleyprocess to limit sales of conflict diamonds, and the CH paper notes that it playedan important role in ending the civil wars in Sierra Leone and Angola. It alsonotes that the same approach has been applied to other natural resources,including timber, and could be applied to yet others, including oil and drugs,many of which originate in areas of civil conflict.

There are other instruments that could also be considered to shortenconflicts, including arms embargoes, diplomacy, and peacekeeping troops, asdiscussed above. The CH paper tested using an econometric approach whetherthere have been any systematic effects of either economic or militaryinterventions on the duration of civil wars. They found that there was no type ofintervention that was systematically effective and thus concluded that theinternational community lacks effective instruments. More detailed studies shouldbe done along these lines, however, considering different types of economic ormilitary intervention and evaluating their effectiveness, as well as consideringother such instruments. For example, it appears that UN peacekeepinginterventions have been successful when they are small missions in terms of

6

funding and numbers of solders but less successful in the case of large missions.Other instruments might include economic sanctions that have worked in somecases despite the often-stated claim that they are never effective (See Hufbauer,Schott, and Elliott, 1997)

7. Instruments for Reducing the Chance of the Resumption of Conflict: Aidand Military Intervention

The CH paper reports that about half of all civil wars are relapses intoconflict during the first decade of peace following a previous conflict. Thus, theyconclude that the diplomatic and military resources of the world communityshould be focused on such situations to try to prevent such a relapse usingwhatever tools at its disposal – economic, political, diplomatic, military, etc.

The CH paper considers only two instruments for reducing the risk ofreversion to conflict in post-conflict situations: aid to promote economic growthand military expenditure. The paper claims that aid is particularly effective inpost-conflict circumstances but that both its scale and its timing have beenflawed. Aid has the same problems in this situation as those discussed above,where the connections between both aid and growth and between growth andreduction of conflict are problematic. In fact, the CH paper finds that on a cost-benefit basis such aid is ineffective, with the net present value of the cost of post-conflict aid, at $12.4 billion, exceeding its benefits of $10.6 billion, although thepaper notes that other global benefits and benefits in terms of poverty reductionhave not been treated. Aid by itself is probably not effective as when it feeds theresources of either side in the conflict, whether directly or through corruption, itonly provides them the wherewithal to resume the conflict. Where it can beeffective is part of a larger package of reform and institution building that ensuresthat it does not stimulate further conflict.

The CH paper mainly treats military intervention in terms of militaryspending rather than the use of foreign or international military forces. It alsoconsiders military intervention only in this situation and not as a way ofpreventing war in the first place or as a way to shorten a conflict even though ithas this potential as well. The paper argues that foreign military intervention islikely to be more effective than domestic military forces. This historical recordalso supports this view. Colonial powers used their military to accomplish bothwar prevention and the shortening of conflicts in their colonies, with armies usedto prevent the outbreak of conflict and to shorten conflicts that did break out. Aclassic example was the French Foreign Legion in the French colonies in Africa.In the post-colonial world the UN peacekeeping forces to some extent serve thisfunction of preventing war outbreak in potential conflicts, of shortening conflictsthat are raging, and of preventing the resumption of conflict in post-conflictsituations, the three situations treated in the paper. In its conclusions, the CHpaper finds that the most effective instrument for preventing civil wars is externalmilitary intervention under Chapter VII of the United Nations Charter to enforce

7

peace in an immediate post-conflict situation and to maintain it over a decadeduring which the risks of reversion to conflict are very high. They find that suchexternal military intervention, if adopted universally in post-conflict situationswould have a massive payoff of over $300 billion, even after assuming that aftertheir withdrawal the situation reverts to what it would have been without themilitary presence.

Along these lines, the CH paper suggests an opportunity for internationalmilitary intervention designed to keep the peace. This opportunity is to combinean external military presence with the government making deep cuts in its militarybudget to obtain a peace dividend. This proposal makes sense as a foreignmilitary presence could be effective in preserving peace, but deep cuts in thegovernment’s military budget are valuable not as a way of obtaining a peacedividend, which is doubtful in any case in the short run. Rather this is a way oflimiting their role in a potential resumption of the civil war and at the same time away of signaling the opposition forces that they are not intending to resume theconflict. The CH paper illustrates an external military presence in the case ofBritish troops in Sierra Leone and suggests that this might be a model for 12other post-conflict situations. (On the peace dividend and its different results inthe long run and the short run, see Intriligator, 1996)

8. Benefits of Reducing the Incidence of Civil War

The CH paper treats the benefits stemming from a reduced incidence ofcivil war at the national, regional, and global level. At the national level theyassert that the benefits stem from avoiding a reduction in the rate of economicgrowth. Surely civil war would lower the growth rate due to the diversion ofresources from production and distribution to war fighting as well as the generallychaotic situation during such a war. Indeed the paper finds that each year of civilwar reduces the growth rate by around 2.2 percent. At the same time, there aredirect costs of lost lives, loss of infrastructure, and many others that go beyondsimply reduced growth. These costs should be considered in any study ofavoiding civil wars.

The CH paper estimates the costs of a typical civil war, or, equivalently, thebenefit stemming from avoiding such a war, as $45 billion in net present value[NPV] terms. This remarkable figure is obtained by their estimate of the loss ofGDP to the country directly affected, amounting to 105 percent of initial GDP, thediversion of spending into the military in the country directly affected, amountingto 18 percent of GDP and a similar diversion in neighbors amounting to 12percent of GDP. The total cost is thus the equivalent of 202 percent of GDP. Theaverage GDP of conflict-affected low-income countries just prior to conflict is$19.7 billion, so the representative conflict cost around $40 billion. To this theyadd health costs of around $5 billion, leading to their estimate of $45 billion asthe cost of a typical civil war in a low-income country. While this cost estimate issubject to many uncertainties, it is a huge amount that dwarfs the cost of

8

potential ways of avoiding such a war, whether through diplomacy, militaryintervention, or other means. The implication is that the world as a whole shouldnot ignore these wars but rather work assiduously to avoid them if at all possible.As the paper properly notes, “Even these lower-bound estimates suggest that theeconomic and social costs of civil war are enormous. As such the issue clearlydeserves international attention. The question is not whether the problem isimportant, but rather whether there are instruments that can be used effectivelyto tackle the problem through international action.”

This conclusion on the importance of avoiding civil wars is reinforced bywhat the authors refer to as the “conflict trap,” namely the high chance that acountry that has had a civil war that ended will have another war. Taking accountof the higher probability of a civil war for a nation suffering such a war, whichamounts to over 40 percent as compared to around 14 percent for a typical low-income country. Discounting back to the beginning of the first conflict, results inan additional cost of around $12 billion, yielding a total cost of $57 billion. Eventhis huge figure does not account for global effects that the authors note areimportant but difficult to quantify. These include drugs, AIDS, and internationalterrorism, with added very large costs. Clearly, there are very high rewards topreventing, shortening, or stopping the resumption of civil wars. This much isclearly demonstrated in the CH paper. What is less convincing are theinstruments that they propose to deal with this issue, whether aid or transparencyin natural resources. These are relatively weak ways of avoiding civil wars, andthere are more potent instruments that could be used, including diplomacy andmilitary intervention.

9. Other Dimensions of the Problem of Avoiding Civil Wars

There are yet other dimensions of the challenge of avoiding civil wars,including those of health, the environment, politics, psychology, internationalrelations, and others as well. It is a mistake to ignore these other dimensions thatcan also have profound economic effects.

As to health, DALYs do not tell the entire story, given the cost of burdeningthe health system, given the fact that the wounded in a civil war must be treated,given the fact that as a result of a civil war people are not able to work or toharvest food, etc. These effects of war via loss of health should be treated. TheCH paper does refer to the deterioration of health as a result of civil war,including forced population movements and the collapse of basic health services,leading to the loss of millions of DALYs, which they compute, using an assumedvalue of $1,000 per DALY as around $5 billion for a typical civil war. This is anunderestimate, however, as it does not account for other health costs involved ina civil war.

As to the environment, wars, including civil wars, have a devastating effecton the environment, which imposes costs on the nation, including loss of

9

agricultural land, loss of potable water, etc. These costs are acute during andafter a war and some are long-term, lasting many years. They can also spill overto neighboring countries.

As to politics, civil wars are sometimes fought to overthrow a brutal dictator,but the result is in some cases to lead to the takeover of the country by anotherdictator and a sequence of civil wars. These political factors also have economicconsequences in terms of lack of an appropriate infrastructure for an economy tofunction, corruption that has the effect of diverting scarce resources, etc.

As to psychology, that is also part of the problem of civil wars and achallenge to stopping them or ending them. Belligerents during such a war argue:"Why stop now?" They will also try to avoid any show of weakness arguingagainst a settlement or even a negotiated end to the conflict by saying that "Wewill look weak." The idea of stopping fighting or not resuming a war will be metwith "We can't give them what they want, so we can't stop fighting," leading tounbridgeable divides, uncontrollable armies, and an inability to stop the war.These factors must be overcome to stop civil wars, and they must be carefullystudied. (See Smith, 1995 and Berdal, 1996)

As to international relations, it is often the case that other nations,particularly neighboring states or great power states play a role in initiating orinhibiting a civil war, and they could play an important role in stopping it. Thereare usually both domestic causes and international factors at work in starting anyparticular civil war, and the international community can play an important role instopping or shortening such wars.

10. Other Conflict Challenges

While there are certainly challenges stemming from civil wars, there areother conflict challenges stemming from international wars, from militaryinterventions, from acts of terrorism, from the burden of spending on arms, fromthe issues of proliferation and arms control, and other related issues that havenot been addressed in the CH paper. These might be addressed in future papersdealing with the global challenge of conflict. Many significant problems ofpotential conflict exist today, and some specific areas that call for further studyand analysis in this area include:

• The dangers of regional conflicts growing from local issues into widerconflagrations

• The potential role of international organizations, major powers, weapons ofmass destruction, deterrence or other mechanisms in avoiding major conflicts

• Arms races, current regional arms races, and the outbreak of war• Arms sales, particularly to nations in unstable regions.• Accidental or inadvertent nuclear war• Proliferation of nuclear weapons and of other weapons of mass

10

destruction, including demand and supply factors• Terrorism, and, in particular, the extremely dangerous potential acquisition

of nuclear weapons or other weapons of mass destruction by terrorist groups.

References

Ali, Tasier M. and Robert O. Matthews, Eds., Civil Wars in Africa: Roots andResolution, Montreal, Ithaca: McGill-Queens University Press, 1999.

Arias, Oscar, “International Code of Conduct on Arms Transfers”Speech at the State of the World Forum, San Francisco, October 5, 1996,http://www.hartford-hwp.com/archives/27a/004.html

Arias, Oscar, “Peace and Justice in the New Century.” Speech presented at theUniversity of San Francisco, September 17, 2002,http://www.usfca.edu/president/arias.html

Berdal, Mats R., Disarmament and demobilisation after civil wars: arms, soldiersand the termination of armed conflicts. New York, NY: Oxford University Pressfor the International Institute for Strategic Studies, Adelphi Paper no. 303, 1996.

[Brahimi] Report of the Panel on United Nations Peace Operations, August 2000.Carter, President Jimmy, Negotiation: The Alternative to Hostility. Macon, GA:Mercer University Press, 1984.

Carter, President Jimmy, Talking Peace. UCLA Burkle Center for InternationalRelations, Bernard Brodie Distinguished Lecture on the Conditions of Peace,2001 http://www.international.ucla.edu/article.asp?parentid=2198Collier, Paul et. al. Breaking the Conflict Trap: Civil War and Development Policy,New York: Oxford University Press, 2003.

Copson, Raymond W, Africa's Wars and Prospects for Peace. Armonk, NY: M. E.Sharpe, 1994.

Hufbauer, Gary C. and Jeffrey J. Schott, and Kimberly Ann Elliott, EconomicSanctions Reconsidered: History and Current Policy, revised edition,Washington, DC: Institute for International Economics, 1997.

Intriligator, Michael D., “The Peace Dividend: Myth or Reality?” Chapter 1 in NilsPetter Gleditsch, Olav Bjerkholt, Odne Cappelen, R.P. Smith and J.P. Dunne,Eds., The Peace Dividend, Amsterdam: Elsevier, 1996.

Klein, Lawrence R. and Kanta Marwah, “Economic Aspects of PeacekeepingOperations,” in Nils Petter Gleditsch, Olav Bjerkholt, Odne Cappelen, R.P. Smithand J.P. Dunne, Eds. The Peace Dividend, Amsterdam: Elsevier, 1996.

11

North, Douglass C., Institutions, Institutional Change, and EconomicPerformance, Cambridge, New York: Cambridge University Press, 1990.

Rotberg, Robert I. et. al., Peacekeeping and Peace Enforcement in Africa:Methods of Conflict Prevention, Cambridge, Mass.: World Peace Foundation,Washington, D.C.: Brookings Institution Press, 2000.

Sandler, Todd, “Fighting Terrorism: What Economics Can Tell Us,”Challenge, Vol. 45, 2002

Schelling, Thomas C., The Strategy of Conflict, Cambridge: Harvard UniversityPress, 1960.

Schelling, Thomas C., Arms and Influence, New Haven: Yale University Press,1966.

Schelling, Thomas C. and Morton H. Halperin, Strategy and Arms Control, NewYork, Twentieth Century Fund, 1961.

Smith, James D. D., Stopping Wars: Defining the Obstacles to Cease-fire.Boulder, CO: Westview Press, 1995.

Stedman, Stephen John, Donald Rothchild, and Elizabeth M. Cousens, EndingCivil Wars: The Implementation of Peace Agreements, Boulder, Colorado: LynneRienner, 2002.